This research project introduces a novel approach to assessing seismic structural damages in local structural elements in reinforced concrete (RC) residential buildings by utilising artificial intelligence (AI) techniques.
To facilitate AI-based damage assessment, this project develops firstly a unified element-level damage state criterion that aligns with the limit state design principles established in current seismic design codes and damage assessment guidelines. The criterion is intended to enable rapid and accurate classification of earthquake-induced damage from image data collected during post-earthquake inspections, so that an AI-based damage assessment model can analyse these images to deduce the governing failure mode at element level, thereby providing actionable evidence to inform engineering decisions on demolition, repair or retrofitting.